回覆列表
-
1 # 機器之心Pro
-
2 # Alice機器學習乾貨鋪
《機器學習》
周志華
如果能把這一本書學會,那麼面試考的基本模型演算法就都不成問題。
《Python Machine Learning》Sebastian Raschka
幾乎每一章都有一個機器學習專案完整的scikit-learn程式碼:
特徵選擇,評估矩陣,模型選擇,評估模型,調優模型
對預處理,降維,超引數調優,模型評估等實際專案中很重要的步驟的講解也很深入,都是一邊講原理,一邊有實戰程式碼。
還有情感分析,預測房價,影象識別等幾個專案。
在應用模型的同時,會講解模型的具體原理,數學公式。
理論和實踐相結合,還有對結果的實際分析。
對於每個模型都會講到其中關鍵的問題,比如 svm 如何選擇核函式,如何選擇重要的特徵等等。
《Hands-On Machine Learning with Scikit-Learn and TensorFlow》Aurélien Géron
這本書非常好,因為只看目錄就會覺得作者的邏輯非常清晰,是能夠系統化掌握機器學習整體知識體系的一本非常不錯的書,機器學習必備。
它的原理可能講的並不深,沒有特別複雜的數學公式,
但是概念比較全面,而且是依照一個完整的專案流程,將核心概念串聯起來,也配有 scikit learn 的程式碼。
它的前一部分是機器學習,用的工具是scikit learn,後一部分是深度學習,工具是Tensorflow,而且今年Tensorflow釋出了2.0版本,這本書也跟著迭代更新,裡面的程式碼全部換成了2.0版本。
-
3 # 532967329
理論方面看:
2、進階級: 《elements of statistical learning》,對數學要求較高,需要惡補線性代數方面的知識。
實戰方面:
1.《機器學習實戰》
2. tensorflow和sklearn的資料和開源專案
其實,更建議您看影片教程:
1、臺灣大學李宏毅的教學影片
2、Andrew NG的影片
3、臺灣大學陳軒田的影片 這些在B站都有
學習是一種理性的投資,每當花費十幾個小時讀完一本書,你就能領略到前人數年積累的經驗。
機器之心之前整理過Swinburne 科技大學的 Jason Brownlee 博士推薦的閱讀數目,適用任何階段的學習者參考。注意這部分推薦的都是英文資料(有的書有中文版),中文資料後面有時間整理補充。
最流行機器學習科普圖書
以下圖書適用於大多數讀者。它們點到了機器學習和資料科學的精華之處,卻沒有使用枯燥的理論或應用細節。這份書單也包括了一些流行的「統計思想」科普書籍。
The Master Algorithm: How the Quest for the Ultimate Learning Machine Will Remake Our World
地址:http://www.amazon.com/dp/0465065708?tag=inspiredalgor-20
Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die
地址:http://www.amazon.com/dp/1119145678?tag=inspiredalgor-20
The Signal and the Noise: Why So Many Predictions Fail–but Some Don"t
地址:http://www.amazon.com/dp/0143125087?tag=inspiredalgor-20
Naked Statistics: Stripping the Dread from the Data
地址:http://www.amazon.com/dp/039334777X?tag=inspiredalgor-20
The Drunkard"s Walk: How Randomness Rules Our Lives
地址:http://www.amazon.com/dp/0307275175?tag=inspiredalgor-20
其中最值得推薦的一本是:《The Signal and the Noise》。
適用於機器學習初學者的書籍
以下列出最適用於初學者的書籍。希望入門的讀者同時也需要參考科普圖書(上一條)以及行業應用圖書(下一條)。
Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking
地址:http://www.amazon.com/dp/1449361323?tag=inspiredalgor-20
Data Smart: Using Data Science to Transform Information into Insight
地址:http://www.amazon.com/dp/111866146X?tag=inspiredalgor-20
Data Mining: Practical Machine Learning Tools and Techniques
地址:http://www.amazon.com/dp/0128042915?tag=inspiredalgor-20
Doing Data Science: Straight Talk from the Frontline
地址:http://www.amazon.com/dp/1449358659?tag=inspiredalgor-20
在這其中最重要的一本是:《Data Mining: Practical Machine Learning Tools and Techniques》。
機器學習入門書籍——高階
以下是適用於希望入門機器學習的本科學生和開發者的書籍,內容包含了機器學習的很多話題,注重如何解決問題,而不是介紹理論。
Machine Learning for Hackers: Case Studies and Algorithms to Get You Started
地址:http://www.amazon.com/dp/B007A0BNP4?tag=inspiredalgor-20
Machine Learning in Action
地址:http://www.amazon.com/dp/1617290181?tag=inspiredalgor-20
Programming Collective Intelligence: Building Smart Web 2.0 Applications
地址:http://www.amazon.com/dp/0596529325?tag=inspiredalgor-20
An Introduction to Statistical Learning: with Applications in R
地址:http://www.amazon.com/dp/1461471370?tag=inspiredalgor-20
Applied Predictive Modeling
地址:http://www.amazon.com/dp/1461468485?tag=inspiredalgor-20
其中最值得推薦的一本是:《An Introduction to Statistical Learning: with Applications in R》
機器學習教材
以下列出了機器學習領域目前最流行的教科書。它們會在研究生課程中出現,包含方法與理論的解讀。
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
地址:http://www.amazon.com/dp/0387848576?tag=inspiredalgor-20
Pattern Recognition and Machine Learning
地址:http://www.amazon.com/dp/0387310738?tag=inspiredalgor-20
Machine Learning: A Probabilistic Perspective
地址:http://www.amazon.com/dp/0262018020?tag=inspiredalgor-20
Learning From Data
地址:http://www.amazon.com/dp/B00YDJC98K?tag=inspiredalgor-20
Machine Learning
地址:http://www.amazon.com/dp/0070428077?tag=inspiredalgor-20
Machine Learning: The Art and Science of Algorithms that Make Sense of Data
地址:http://www.amazon.com/dp/1107422221?tag=inspiredalgor-20
Foundations of Machine Learning
地址:http://www.amazon.com/dp/026201825X?tag=inspiredalgor-20
其中的重點是:《The Elements of Statistical Learning: Data Mining, Inference, and Prediction》
機器學習圖書——按主題分
有關 R 語言在機器學習中如何應用的圖書。
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
地址:http://www.amazon.com/dp/0387848576?tag=inspiredalgor-20
Pattern Recognition and Machine Learning
地址:http://www.amazon.com/dp/0387310738?tag=inspiredalgor-20
Machine Learning: A Probabilistic Perspective
地址:http://www.amazon.com/dp/0262018020?tag=inspiredalgor-20
Learning From Data
地址:http://www.amazon.com/dp/B00YDJC98K?tag=inspiredalgor-20
Machine Learning
地址:http://www.amazon.com/dp/0070428077?tag=inspiredalgor-20
Machine Learning: The Art and Science of Algorithms that Make Sense of Data
地址:http://www.amazon.com/dp/1107422221?tag=inspiredalgor-20
Foundations of Machine Learning
地址:http://www.amazon.com/dp/026201825X?tag=inspiredalgor-20
這方面的首選圖書是:《The Elements of Statistical Learning: Data Mining, Inference, and Prediction》。
Python 機器學習
以下列出 Python 機器學習熱門書籍
Python Machine Learning
地址:http://www.amazon.com/dp/1783555130?tag=inspiredalgor-20
Data Science from Scratch: First Principles with Python
地址:http://www.amazon.com/dp/149190142X?tag=inspiredalgor-20
Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques for Building Intelligent Systems
地址:http://www.amazon.com/dp/1491962291?tag=inspiredalgor-20
Introduction to Machine Learning with Python: A Guide for Data Scientists
地址:http://www.amazon.com/dp/1449369413?tag=inspiredalgor-20
Vital Introduction to Machine Learning with Python: Best Practices to Improve and Optimize Machine Learning Systems and Algorithms
地址:http://www.amazon.com/dp/B01N4FUDSE?tag=inspiredalgor-20
Machine Learning in Python: Essential Techniques for Predictive Analysis
地址:http://www.amazon.com/dp/1118961749?tag=inspiredalgor-20
Python Data Science Handbook: Essential Tools for Working with Data
地址:http://www.amazon.com/dp/1491912057?tag=inspiredalgor-20
Introducing Data Science: Big Data, Machine Learning, and more, using Python tools 地址:http://www.amazon.com/dp/1633430030?tag=inspiredalgor-20
Real-World Machine Learning
地址:http://www.amazon.com/dp/1617291927?tag=inspiredalgor-20
最值得注意的當然是《Python 機器學習》了。
深度學習
注意:深度學習的圖書目前還比較稀缺,以下這份列表只能保證數量,而不是質量。
Deep Learning
地址:http://www.amazon.com/dp/0262035618?tag=inspiredalgor-20
Deep Learning: A Practitioner"s Approach
地址:http://www.amazon.com/dp/1491914254?tag=inspiredalgor-20
Fundamentals of Deep Learning: Designing Next-Generation Machine Intelligence Algorithms
地址:http://www.amazon.com/dp/1491925612?tag=inspiredalgor-20
Learning TensorFlow: A guide to building deep learning systems
地址:http://www.amazon.com/dp/1491978511?tag=inspiredalgor-20
Machine Learning with TensorFlow
地址:http://www.amazon.com/dp/1617293873?tag=inspiredalgor-20
TensorFlow Machine Learning Cookbook
地址:http://www.amazon.com/dp/1786462168?tag=inspiredalgor-20
Getting Started with TensorFlow
地址:http://www.amazon.com/dp/1786468573?tag=inspiredalgor-20
TensorFlow for Machine Intelligence: A Hands-On Introduction to Learning Algorithms
地址:http://www.amazon.com/dp/1939902452?tag=inspiredalgor-20
其中最重要的一本書當然是:Yoshua Bengio 和 Ian Goodfellow 所著的《Deep Learning》(此書中文版網上已有)。
時序序列預測
目前時序序列預測在實際應用中主要是由 R 語言的平臺所主導。
Time Series Analysis: Forecasting and Control
地址:http://www.amazon.com/dp/1118675029?tag=inspiredalgor-20
Practical Time Series Forecasting with R: A Hands-On Guide
地址:http://www.amazon.com/dp/0997847913?tag=inspiredalgor-20
Introduction to Time Series and Forecasting
地址:http://www.amazon.com/dp/3319298526?tag=inspiredalgor-20
Forecasting:principles and practice
地址:http://www.amazon.com/dp/0987507109?tag=inspiredalgor-20
最優質的入門介紹書籍是 Forecasting:principles and practice。
時序序列最優質的教科書是 Time Series Analysis: Forecasting and Control。